Quality-Trigger False Intervention Rate (QTFIR)

GMI-MFG-028ManufacturingPilot-Ready

Definition

Measures the percentage of quality-linked condition monitoring-triggered maintenance events where post-maintenance analysis confirms the maintenance was unnecessary.

The Measurement Gap

Quality-linked condition monitoring represents an advancement over traditional time-based maintenance because it triggers maintenance when asset condition indicators suggest quality degradation is imminent. However, the threshold settings that govern when the CM system triggers a maintenance intervention are typically set conservatively to avoid quality escapes. Conservative thresholds mean the system triggers maintenance earlier than necessary, creating production interruptions that were not needed because the quality metric would have remained within specification through the next scheduled maintenance window.

The measurement gap is the absence of any feedback loop from post-maintenance analysis back to the CM trigger threshold. Plants track whether maintenance was performed on time and whether quality is in specification, but nobody tracks whether the maintenance was necessary given the actual quality trajectory at the time of the trigger. QTFIR closes this loop by assessing, after each quality-CM-triggered maintenance event, whether the quality metric was actually trending toward the specification limit or whether it was stable and the trigger was a false alarm. This feedback enables systematic threshold optimization rather than the current practice of setting thresholds based on engineering judgment and never revisiting them.

Formula

QTFIR divides the count of unnecessary quality-CM-triggered maintenance events by the total count of quality-CM-triggered events, expressed as a percentage.

QTFIR = Interventions_unnecessary / Interventions_QCM_total times 100

Where Interventions_unnecessary is the count of quality-CM-triggered maintenance events where post-hoc review (quality trend analysis, process capability assessment) indicates the quality metric would have remained within specification through the next scheduled maintenance window. Interventions_QCM_total is all maintenance events triggered by the quality-linked CM system in the measurement period.

A QTFIR of 0% means every quality-CM trigger was necessary, and the asset genuinely needed maintenance to prevent a quality excursion. A QTFIR of 100% means every trigger was a false alarm and the CM threshold is set far too aggressively. Higher values indicate more wasted maintenance effort.

The post-hoc assessment requires a standard criterion to determine necessity. A recommended standard: if the quality metric was more than 20% of the tolerance band away from the specification limit and the trend was not accelerating at the time of the trigger, classify the intervention as unnecessary. This criterion is judgment-based and should be reviewed by quality engineering during the first two measurement periods to calibrate the threshold.

QTFIR is distinct from the In-Process False Positive Rate (IFPR) in the broader manufacturing catalog, which measures quality prediction AI false positives. The trigger direction is reversed: IFPR asks whether the AI flagged a quality problem that was not there. QTFIR asks whether the CM system triggered maintenance that the asset did not need from a quality perspective. Minimum sample is 15 quality-CM-triggered maintenance events per measurement period.

Proposed formula, pending validation pilots.

At a Glance

Unit
Percentage
Trend Direction
Lower
Reporting Frequency
Quarterly
Category
Quality-Linked Maintenance Calibration
← Product-Conditional Confidence Calibration Index (PCCCI)Certification Verification at Assignment Rate (CVAR-MFG) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment